EP3230846B1 - Exécution d'une opération d'écriture atomique sur une pluralité de dispositifs de stockage - Google Patents

Exécution d'une opération d'écriture atomique sur une pluralité de dispositifs de stockage Download PDF

Info

Publication number
EP3230846B1
EP3230846B1 EP15867697.3A EP15867697A EP3230846B1 EP 3230846 B1 EP3230846 B1 EP 3230846B1 EP 15867697 A EP15867697 A EP 15867697A EP 3230846 B1 EP3230846 B1 EP 3230846B1
Authority
EP
European Patent Office
Prior art keywords
write
data
storage devices
command
storage
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP15867697.3A
Other languages
German (de)
English (en)
Other versions
EP3230846A1 (fr
EP3230846A4 (fr
Inventor
Dan J. WILLIAMS
Bryan E. Veal
Annie Foong
Sanjeev N. Trika
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Intel Corp
Original Assignee
Intel Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Intel Corp filed Critical Intel Corp
Publication of EP3230846A1 publication Critical patent/EP3230846A1/fr
Publication of EP3230846A4 publication Critical patent/EP3230846A4/fr
Application granted granted Critical
Publication of EP3230846B1 publication Critical patent/EP3230846B1/fr
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1402—Saving, restoring, recovering or retrying
    • G06F11/1474—Saving, restoring, recovering or retrying in transactions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/0614—Improving the reliability of storage systems
    • G06F3/0619—Improving the reliability of storage systems in relation to data integrity, e.g. data losses, bit errors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0638—Organizing or formatting or addressing of data
    • G06F3/064—Management of blocks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
    • G06F3/065—Replication mechanisms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/0671—In-line storage system
    • G06F3/0683—Plurality of storage devices
    • G06F3/0689—Disk arrays, e.g. RAID, JBOD

Definitions

  • Embodiments described herein generally relate to a host system storing data across multiple storage devices by performing an atomic write operation across the storage devices.
  • Redundant Array of Independent Disk controllers stripe data to multiple storage drives and calculate parity data from the data written to the drives in the RAID stripes (typically by XOR'ing the striped data) to write to a parity disk.
  • RAID Redundant Array of Independent Disk
  • One potential problem with RAID implementations is known as a RAID write hole. A write hole occurs when data in a RAID stripe is updated, but the parity data is not updated so that all disks XOR to zero in order to reconstruct data when a disk fails. The problem is that there is no way to update two or more disks atomically, so RAID stripes can become damaged during a crash or power outage.
  • RAID controllers maintain non-volatile storage devices, also known as NVRAMs, to backup the write data until the parity data can be written.
  • NVRAMs non-volatile storage devices
  • adding NVRAMs increases the cost of RAID controllers and RAID solutions.
  • RAID software implementations have not been known to successfully address the write hole problem without having significant performance issues.
  • US2012/030408 discusses system and method for atomic storage operations, where a virtual storage layer (VSL) for a non-volatile storage device presents a logical address space of a non-volatile storage device to storage clients.
  • Storage metadata assigns logical identifiers in the logical address space to physical storage locations on the non-volatile storage device.
  • Data is stored on the non-volatile storage device in a sequential log-based format.
  • Data on the non-volatile storage device comprises an event log of the storage operations performed on the non-volatile storage device.
  • the VSL presents an interface for requesting atomic storage operations. Previous versions of data overwritten by the atomic storage device are maintained until the atomic storage operation is successfully completed.
  • Data pertaining to a failed atomic storage operation may be identified using a persistent metadata flag stored with the data on the non-volatile storage device. Data pertaining to failed or incomplete atomic storage requests may be invalidated and removed from the non-volatile storage device.
  • US2013/198446 discusses a storage system for atomic write of one or more commands.
  • WO02/071230 discusses a method for utilizing parity caching and parity logging while closing the RAID5 write hole.
  • Described embodiments provide techniques to implement a RAID or other storage controller that protects against write holes and silent data corruption while avoiding the use of NVRAM devices by offloading the storing of partially written data to the storage devices themselves.
  • the storage or RAID controller coordinates with the distributed storage devices to ensure atomic updates to the storage devices without the need of a separate write back cache or a commit log device to store the partially written data.
  • Described embodiments provide a device distributed solution for updating multiple discontiguous blocks of data across multiple devices in an atomic manner.
  • the storage controller which may be implemented in software in the host system, can assert that the entire operation completes in whole or not at all by offloading the storing of the partially or fully written data at the storage devices.
  • their read and write rates can approximate that of an NVRAM.
  • described embodiments do not need an NVRAM to backup the partially written data, which is offloaded to the storage devices, SSDs. In this way, when the storage devices are fast access devices, such as SSDs, there will be no bottleneck introduced by the NVRAM because the SSDs will be used in atomic write operations to store the partially written data and previous version of the data.
  • Storage devices implementing the described embodiments provide a mechanism for a storage controller, such as a software RAID controller, to register the intent in the storage devices to backup a previous version of the data.
  • This "intent” indicates that the storage drives will provide the data that is needed for recovery in the event of an interruption to the writes.
  • the storage devices While an intent entry is open for a write operation, the storage devices maintain an "old” and “new” copy of the listed data blocks. Once the write completes the intent entry and old data in each device can be discarded. However, if the write is interrupted, there is sufficient distributed metadata across all the participating storage devices and the storage controller for a RAID implementation, multi-device file system, or database to either roll the transaction forward to completion, or rollback to a previous state/data. Consumers of the described embodiment storage system experience the write request completing in its entirety or not at all.
  • Embodiments relate to storage devices electronic assemblies. Embodiments include both devices and methods for forming electronic assemblies.
  • FIG. 1 illustrates an embodiment of a storage environment having a host system 100 which stores data, such as user data that may be communicated from other attached computer systems and devices, to multiple storage devices 102 1 , 102 2 ...102 n .
  • the host system 100 includes an operating system 104 including a storage controller driver 106 to manage the storage of data in the storage devices 102 1 , 102 2 ...102 n .
  • the storage controller driver 106 maintains write operation status information 200 having information on the data distributed to the storage devices 102 1 , 102 2 ...102 n for storage and a write cache 110 that buffers the data sent to the storage devices 102 1 , 102 2 ...102 n until a storage device confirms that the data has been written.
  • the storage controller driver 106 may implement Redundant Array of Independent Disk (RAID) algorithms stripe to portions of a data unit to a plurality of the storage devices storage devices 102 1 , 102 2 ...102 n , calculate parity data for the striped data units, and store the parity data in one of the storage devices 102 1 , 102 2 ...102 n .
  • the parity data may comprise XOR parity data to perform parity error recovery if one the storage devices 102 1 , 102 2 ...102 n having the data fails.
  • the storage controller driver 106 may implement different RAID algorithms, including interleaving, etc.
  • the storage controller driver 106 comprises a software component of the operating system 104.
  • the storage controller driver 106 may be implemented in a hardware device or separate hardware controller or card which may be connected to the host system 100 through an internal bus.
  • Each storage device 102 1 , 102 2 ...102 n includes a controller 112 that manages the writing of data to one or more data storage units 114, including communicating and receiving read and write commands from the host system 100.
  • the controller 112 maintains write operation status 300 having information on the status of write requests sent by the host system 100.
  • the controller 112 may also include a queue to queue read and write requests from the host system 100.
  • the storage devices storage device 102 1 , 102 2 ...102 n may comprise a solid state drive (SSD), flash drive, etc.
  • the data storage units 114 may comprise NAND dies of flash memory cells.
  • the NAND dies may comprise a multilevel cell (MLC) NAND flash memory that in each cell records two bit values, a lower bit value and an upper bit value.
  • the NAND dies may comprise single level cell (SLC) memories or three bit per cell (TLC) memories.
  • the storage units 114 may also comprise, but not limited to, MLC NAND flash memory, ferroelectric random-access memory (FeTRAM), nanowire-based non-volatile memory, three-dimensional (3D) crosspoint memory such as phase change memory (PCM), memory that incorporates memristor technology, Magnetoresistive random-access memory (MRAM), Spin Transfer Torque (STT)-MRAM, a single level cell (SLC) Flash memory, energy source backed (battery / super-capacitor) DRAM, and other electrically erasable programmable read only memory (EEPROM) type devices.
  • the storage devices 102 1 , 102 2 ...102 n may comprise other types of storage devices, such as a hard disk drives, etc.
  • the data storage unit 114 may comprise the magnetic disk.
  • the host 100 may communicate with the storage devices 102 1 , 102 2 ...102 n over an interface 116 such as a bus interface.
  • the host 100 and storage devices 102 1 , 102 2 ...102 n may be located in a same enclosure or separate enclosures.
  • the host system 100 may communicate with the storage devices storage devices 102 1 , 102 2 ...102 n over a network.
  • the host 100 and storage devices 102 1 , 102 2 ...102 n may include Input/Output (I/O) interfaces to allow communication therebetween, such as a Serial Advanced Technology Attachment (SATA) interface and include a SATA compliant bus coupling the host 100 and the storage device 102.
  • I/O interfaces and bus interconnects may be used, such as Serial Attached Small Computer System Interface (SCSI) (or simply SAS), Peripheral Component Interconnect Express (PCIe), etc.
  • FIG. 2 illustrates an embodiment of a write operation status entry 200 i the storage controller driver 106 maintains for each write operation opened with respect to one of the storage devices storage devices 102 1 , 102 2 ...102 n .
  • Each entry 200 i may have a write operation tag 200 providing a unique identifier of a write operation, which may have tracks or blocks of data striped across multiple storage devices 102 1 , 102 2 ...102 n ; a storage device identifier (ID) identifying the storage device 102 1 , 102 2 ...102 n to which the write operation identified by the tag 200 is directed; and a state 206 of the write operation.
  • ID storage device identifier
  • a write operation involves writing portions of a data unit to multiple of the storage devices 102 1 , 102 2 ...102 n , then there may be multiple write status entries 200 i for each write operation, each of these entries providing the status for the identified write operation with respect to one of the storage devices 102 1 , 102 2 ...102 n having received data for the write operation tag 202.
  • the state 206 may indicate a state of the write operation with respect to one of the storage devices 102 1 , 102 2 ...102 n , including, but not limited:
  • FIG. 3 illustrates an embodiment of a write operation status entry 300 i that each of the storage devices 102 1 , 102 2 ...102 n maintain for each write operation opened by the storage controller driver 106.
  • Each entry 300 i may have a write operation tag 300 providing a unique identifier of a write operation, which may have tracks or blocks of data striped across multiple storage devices 102 1 , 102 2 ...102 n ; an intent logged flag 304 indicating the intent of the storage device to maintain a copy of a previous version of the data that is updated; and a state 306 of the write operation, which may comprise one of the states described above.
  • the storage controller driver 106 may not maintain explicit write operation status 200 information during the operations. Instead, the storage devices 102 1 , 102 2 ...102 n maintain the status 300, and the storage controller driver 106 would obtain the status from the storage devices 102 1 , 102 2 ...102 n when needed to handle an interruption to write operations.
  • the storage controller driver 106 may submit the below commands to the storage devices 102 1 , 102 2 ...102 n as part of a write operation to write portions of a data unit or extent to multiple storage devices 102 1 , 102 2 ...102 n :
  • FIG. 4 illustrates an embodiment of a state diagram 400 describing the different states at the storage devices 102 1 , 102 2 ...102 n .
  • the old data state 401 shows the state of the data before it is updated.
  • the state of the receiving storage device 102 1 , 102 2 ...102 n transitions 402 to an intent logged state 403 where the storage device 102 1 , 102 2 ...102 n saves the previous version of the data before writing the new data.
  • the storage device 102 1 , 102 2 ...102 n may write the new data by performing a write-in-place write or use indirection to perform a copy on-write, to copy the old data upon writing.
  • the state of the receiving storage device 102 1 , 102 2 ...102 n transitions 404 to a dirty state 405 when a particular write for one of the portions of the data unit completes, but before all the writes for the storage device have completed.
  • the state transitions 407 and 408, respectively, to the written state 409.
  • the storage device 102 1 , 102 2 ...102 n transitions 410 to the new data state 411.
  • the storage device 102 1 , 102 2 ...102 n discards or the logged previous version of the data and other information of the write operation (tag), such as the write status entry 300 i for the write operation.
  • the storage space having the logged previous version of the data and other related information may be marked as free space and available for garbage collection or reuse.
  • the state diagram embodiment of FIG. 4 provides atomicity because the write operation does not complete at any of the storage devices 102 1 , 102 2 ...102 n until all the storage devices have confirmed that their data has been written, either in a single write command or multiple write commands. Until the point at which the atomic write operation completes, the storage devices 102 1 , 102 2 ...102 n maintain partial and full written data as well as a previous version of the data to allow the storage controller driver 106 to roll back the data for all the storage devices to the previous version of the data before the update or the write data of the write operation.
  • the storage devices 102 1 , 102 2 ...102 n maintain sufficient information to allow the calculation of parity data before the atomic write operation completes.
  • the implementation of this atomic write operation between the storage controller driver 106 and storage devices 102 1 , 102 2 ...102 n is described in detail below.
  • FIGs. 5a and 5b illustrate an embodiment of operations performed by the storage controller driver 106 to manage the atomic write operation having write data, e.g., data units, tracks, extents, to write or stripe across multiple storage devices 102 1 , 102 2 ...102 n .
  • the storage controller 106 determines (at block 501) a tag to use to uniquely identify the write operation.
  • the storage controller driver 106 sends (at block 503) an OPEN command to each of the storage devices 102 1 , 102 2 ...102 n to receive data for the write operation to instruct the storage devices 102 1 , 102 2 ...102 n to start intent logging to store a previous version of the data to be updated.
  • Sending the OPEN command results in the storage controller driver 106 creating a write status entry 200i for each storage device 102 1 , 102 2 ...102 n to which data for the write operation will be sent having a status of IDLE. If the write operation will involve multiple write commands for portions of the data for each storage device, then the storage drive controller 106 may provide a write status entry 200 i having a status of IDLE for each of the write commands writing a portion of the data in order to provide separate status for each of the write commands for the portions of the data for the storage device 102 1 , 102 2 ...102 n .
  • the storage controller driver 106 Upon receiving (at block 505) acknowledgment of the OPEN command from one of the storage devices 102 1 , 102 2 ...102 n for the write operation, the storage controller driver 106 updates (at block 506) the status 206 for the write operation tag and storage device 102 1 , 102 2 ...102 n to indicate OPEN. If (at block 507) all of the storage devices 102 1 , 102 2 ...102 n do not have the OPEN state, then the storage controller driver 106 returns to block 505 to wait for acknowledgment of the OPEN command from other of the storage devices 102 1 , 102 2 ...102 n to insure that all storage devices 102 1 , 102 2 ...102 n acknowledge they will engage in intent logging.
  • the storage controller driver 106 determines (at block 508) data to send to each of the storage devices 102 1 , 102 2 ...102 n for the write operation.
  • This operation may involve determining portions of a data unit to write across multiple storage devices. In RAID implementations, this operation may determine data to stripe to each of the storage devices 102 1 , 102 2 ...102 n and a calculated XOR parity data for one or more of the storage devices 102 1 , 102 2 ...102 n storing the XOR parity data. Further, this determination may further break the portion of data to write for each of the storage devices into multiple write commands to write each portion of the data for each storage device 102 1 , 102 2 ...102 n .
  • the storage controller driver 106 sends (at block 509) a WRITE (or WRITE WITH TAG) command to each of the storage devices to write the indicated write data determined for the storage device.
  • the state 206 is updated (at block 510) for the write operation tag and each of the storage devices 102 1 , 102 2 ...102 n receiving one or more write commands to indicate that the write was sent.
  • the OPEN state for a storage device having received the WRITE command would indicate the write was sent but has not completed.
  • Control then proceeds to block 511 in FIG. 5b where the storage controller driver 106 receives (at block 511) acknowledgment that the write completed from one of the storage devices for the write operation tag.
  • the state for the write operation tag and acknowledging storage device is updated (at block 512) to indicate that the write completed, which may be the WRITTEN state.
  • the storage controller driver 106 may then clear (at block 513) the data for that completed write operation in the write cache 110. At this point, the copy of the write data is maintained at the storage device 102 1 , 102 2 ...102 n and the storage controller driver 106 does not need to maintain partially written data.
  • the storage controller driver 106 may discard the portion of the data for the write command, but retain portions of the data to write for the storage device for write commands for which acknowledgment has not been received. If (at block 514) there are some storage devices 102 1 , 102 2 ...102 n receiving the write command that have not acknowledged write complete state, e.g., the WRITTEN state, then control proceeds back to block 511 to wait for further write complete acknowledgment from all the storage devices 102 1 , 102 2 ...102 n . For a write operation involving sending multiple write commands to each of the storage devices 102 1 , 102 2 ...102 n , control proceeds back to block 511 if not all of the write commands for portions of the data have completed.
  • the storage controller driver 106 sends (at block 515) a CLOSE command to each of the storage devices 102 1 , 102 2 ...102 n identified in the write status entries 200 i for the write operation tag.
  • the storage controller driver 106 Upon receiving (at block 516) acknowledgment that the CLOSE command completed from one of the storage devices 102 1 , 102 2 ...102 n for the write operation, the storage controller driver 106 updates (at block 517) the state 206 in the write status entry 200 i for the write operation tag 202 and the acknowledging storage device 204 to indicate the CLOSE status.
  • the operations at blocks 518 and 519 are not performed, and the CLOSE is indicated for each storage device 102 1 , 102 2 ...102 n asynchronously when the acknowledgment is received for the first time, because all devices have written the data by the time the first CLOSE state acknowledgment is received.
  • the storage controller driver 106 does not maintain the state information based on the acknowledgments received from the storage devices 102 1 , 102 2 ...102 n , and only obtains such state information when needed when querying the storage devices 102 1 , 102 2 ...102 n for the state information when there is an interruption to write operations or other failure.
  • the storage controller driver 106 does not need to maintain all the write data and previous version of the write data until the entire write operation is confirmed as completed. Instead, once data is written to the storage device 102 1 , 102 2 ...102 n , the data for that storage device 102 1 , 102 2 ...102 n may be discarded because the atomic write is assured that each of the storage devices 102 1 , 102 2 ...102 n maintain a previous copy of the data and the current write data to allow for rolling back or for completing the write.
  • parity data may comprise erasure code parity data, such as XOR parity and other parity data, that can be used for data recovery if one of the storage devices fails.
  • FIGs. 6-9 illustrate operations implemented in the controller 112 of each of the storage devices 102 1 , 102 2 ...102 n to process the commands from the storage controller driver 106 to implement the atomic write operation.
  • FIG. 6 illustrates an embodiment of operations performed by the controller 112 to process the OPEN command from the storage controller driver 106.
  • the storage device controller 112 creates (at block 601) a write status entry 300 i identifying the write operation tag 302, indicating the intent logged 304 state indicating that a previous version of data to update as part of the identified write operation tag 302 will be logged and saved, and indicating the state 306 as OPEN.
  • the controller 112 returns (at block 602) OPEN complete state to the host 100.
  • a storage device 102 1 , 102 2 ...102 n having an OPEN state acknowledges an intent to save the previous version of the updated data and that any sent writes have not been received or have been received but not completed.
  • FIG. 7 illustrates an embodiment of operations performed by the storage device controller 112 to process a WRITE command for a write operation tag, such as the WRITE WITH TAG.
  • the controller 112 Upon receiving (at block 700) a write command for a write operation (tag) having write data for a logical or physical address addressing a location in the data storage unit(s) 114, the controller 112 saves (at block 701) a copy of the data in the data storage unit 114 at the storage address as the previous version of the data to update.
  • the controller 112 can remember where the old data is located and write the new data to a new storage area, and an indirection table allows the location of the old data to be reused.
  • the write data in the write command is written (at block 702) to the storage address.
  • a write may be deemed completed upon writing the data to the actual storage unit 114, such as the case with flash memory, SSDs, etc. or upon buffering the write data in the storage device 102 1 , 102 2 ...102 n in a buffer separate from the data storage unit 114, where the buffered data is eventually destaged to the data storage unit 114, such as the case with a hard disk drive.
  • the controller 112 indicates (at block 703) the state 306 of the write operation as WRITTEN in the write status entry 300 I for the write operation 302. Complete is returned (at block 704) to the host 100.
  • the operations of FIG. 7 may be performed when there is just one write command sent for the write operation to each storage device 102 1 , 102 2 ...102 n .
  • FIG. 8 illustrates an embodiment of operations performed by the storage device controller 112 to process a plurality of WRITE commands, such as the WRITE WITH TAG, to write portions of the data for one storage device and write operation tag.
  • a plurality of write commands also referred to as sub-write commands
  • the controller 112 performs a loop of operations at blocks 801 through 805 for each WRITE.
  • a copy of the data in the data storage unit 114 at the storage address is saved as the previous version of the data to update.
  • the controller 112 can remember where the old data is located and write the new data to a new storage area, and an indirection table allows the location of the old data to be reused.
  • the write data in the WRITE command is written (at block 803) to the storage address.
  • the controller 112 indicates (at block 804) the state 306 of the write operation and the partial data written as DIRTY in the write status entry 300 I for the write operation 302.
  • the controller 112 return write complete for the write commands and the write operation (tag) to the host 100.
  • write complete or WRITTEN state is only returned after performing all the write command to write all the portions of data for the data unit.
  • the controller 112 may return dirty status when each write command is completed, indicating that portion of the write data completed.
  • FIG. 9 illustrates an embodiment of operations performed by the controller 112 to stripe the received write data across a plurality of data units 114, including a parity data unit 114, when the data units of the storage device 102 1 , 102 2 ...102 n comprise independently addressable storage devices.
  • the data units may 114 comprise independent storage devices, such as flash devices, SSDs, hard disk drives, etc.
  • the storage device comprises an SSD
  • the data units 114 may comprise dies of memory cells in the storage device 102 1 , 102 2 ...102 n , such as NAND dies for an SSD storage device.
  • the controller 112 may comprise a RAID controller having a write cache to cache all the data to stripe across the data units 114.
  • the controller 112 may implement the logic of the storage controller driver 106 to perform atomic write operations as described with respect to the storage controller driver 106.
  • the controller 112 Upon processing (at block 900) a write command for a write operation (tag) having write data for a storage address, the controller 112 saves (at block 901) a copy of the data in the data storage units 114 at the storage address as the previous version of the data in a write cache.
  • the controller 112 can remember where the old data is located and write the new data to a new storage area, and an indirection table allows the location of the old data to be reused.
  • the controller 112 determines (at block 902) stripe units for the write data to stripe across data storage units in the storage device and calculates (at block 903) erasure code parity data, such as, but not limited to, XOR parity data, for the stripe units to write to a parity data storage unit of the storage units 114.
  • the controller 112 may use a RAID algorithm to determine the stripe units and parity data according to different RAID levels.
  • the controller 112 writes (at block 904) the stripe units and parity data to the data storage units 904 and returns (at block 905) to the host 110 upon completing the stripping of the data from the write command to the storage units 114. Parity data may be calculated using error correction codes or other parity algorithms known in the art.
  • FIG. 10 illustrates an embodiment of operations performed by the storage device controller 112 to process a CLOSE command from the storage controller driver 106.
  • the controller 112 Upon receiving (at block 1000) the CLOSE command for a write operation (tag), the controller 112 discards (at block 1002) the previous versions of the data for the write operation and the write status information for the write operation, including the write status entry 300 i .
  • CLOSE complete is returned (at block 1004) for the write operation (tag) to the host system 100.
  • FIGs. 11-13 illustrate embodiments of operations performed by the storage controller driver 106 to recover from a failure, such as a power-on failure or other event that results in an interruption of the write operations.
  • the storage controller driver 106 upon detecting (at block 1100) an interruption to the write operations, the storage controller driver 106 sends (at block 1101) a status inquiry to each of the storage devices 102 1 , 102 2 ...102 n not having the WRITTEN state 206 Upon receiving (at block 1102) the state, the storage controller driver 106 updates (at block 1103) the write status entry 200 i for the write operation tag 202 and storage device 204 to indicate the state 206 corresponding to the status received from the storage device, such as WRITTEN, OPEN, CLOSED, etc.
  • the returned state may indicate DIRTY, such that part of the data was written, but not all of the portions of the data sent with different WRITE commands. If (at block 1104) status is received for all of the storage devices 102 1 , 102 2 ...102 n for one of the write operation (tags), then the storage controller driver 106 initiates (at block 1105) a recovery operation to return the storage devices 102 1 , 102 2 ...102 n involved in the write operation to the previous version of the data (roll back) or the version of the write data (roll forward).
  • the recovery operation may involve making the data in the storage device consistent by using the read with tag command to regenerate data that was in flight at the time of the power loss that can be recovered from the storage device 102 1 , 102 2 ...102 n though the read command.
  • FIGs. 12 and 13 illustrate embodiments of operations for different types of recovery operations performed at block 1105 in FIG. 11 .
  • FIG. 12 illustrates an embodiment of operations performed by the storage driver controller 106 to perform the recovery operation to revert back to the previous version of the data before the write operation. If there is a recovery operation, the operations of FIG. 12 may be performed for each write operation tag in the write operation status 200.
  • the storage controller driver 106 determines (at block 1201) whether the state of the write operation is an interrupted write, which occurs when some but not all of the storage devices 102 1 , 102 2 ...102 n in the write status entries 200 i for the write operation tag are in the WRITTEN state. This may occur when some entries 200 i indicate the WRITTEN state and others indicate the OPEN or DIRTY state or any other state indicating a write may not have completed.
  • the storage controller driver 106 determines (at block 1202) whether the state of the write operation tag is an interrupted open, which occurs when at least one of the storage devices 102 1 , 102 2 ...102 n are in an OPEN state and at least one is in the IDLE state. If so, then the recovery action comprises sending (at block 1203) a CLOSE command to all of the storage devices 102 1 , 102 2 ...102 n . Otherwise, if (at block 1202) all are in the IDLE state, then control ends without any recovery action taken.
  • the storage controller driver 106 sends (at block 1204) a REVERT command to any storage device 102 1 , 102 2 ...102 n indicated in write status entries 200 i as having a WRITTEN or DIRTY state to cause the storage device 102 1 , 102 2 ...102 n to roll back the updated data with the previous version of the data.
  • FIG. 13 illustrates an alternative embodiment of operations performed by the storage driver controller 106 to perform the recovery operation to complete the write operation at all of the storage devices 102 1 , 102 2 ...102 n having received the write command for the write operation. If there is a recovery operation, the operations of FIG. 13 may be performed for each write operation tag in the write operation status 200.
  • the storage controller driver 106 Upon initiating the recovery operation (at block 1300), the storage controller driver 106 performs the operations at blocks 1301, 1302, and 1303 which are the same operations described with respect to blocks 1201, 1202, and 1203 in FIG. 12 to determine if the write operation has the interrupted write state or the interrupted OPEN state.
  • the storage controller driver 106 Upon determining (at block 1301) that the state is the interrupted write state, for embodiments having parity data, if (at block 1306) the storage device(s) having parity data is not in the WRITTEN state, then the storage controller driver 106 reads (at block 1304) the written write data from any storage devices having entries 200 i indicating the DIRTY or WRITTEN state 206. The storage controller driver 106 calculates (at block 1305) parity data from any of the write data read from the storage devices 102 1 , 102 2 ...102 n and write data in the write cache 110, for uncompleted writes. The calculated parity data is written (at block 1306) to the storage device(s) having parity.
  • the storage controller driver 106 After completing rewriting the parity (at block 1306) or if (at block 1303) the storage device(s) having parity data are in the WRITTEN state, then for each of the storage devices 102 1 , 102 2 ...102 n identified in status entries 200 i for the write operation tag as in the OPEN state (write did not complete), then the storage controller driver 106 resends (at block 1307) the WRITE command to write the previously sent write data from the write cache 110. In embodiments where there is no parity data, then after determining the interrupted write state for the write operation (at block 130), control would proceed directly to block 1307.
  • the operation at block 1307 would be performed to rewrite data for storage devices that did not complete the write.
  • the storage device 102 1 , 102 2 ...102 n may not return intermediary DIRTY acknowledgments upon completing the write, but may wait until all the WRITE commands or sub-write commands complete and then send a write complete for all the write commands used for the writ operation.
  • the DIRTY state is only returned in response to the status inquiry during error recovery.
  • the storage device 102 1 , 102 2 ...102 n may return DIRTY in response to completing one of a plurality of write commands for a write operation.
  • FIG. 14 illustrates an embodiment of operations performed by the controller 112 of the storage devices 102 1 , 102 2 ...102 n in response to receiving the status inquiry form the storage controller driver 106.
  • the storage device 102 1 , 102 2 ...102 n will complete the pending commands for the write operations, such as OPEN, WRITE, CLOSE, REVERT.
  • the controller 112 sends (at block 1403) the state information for all the write operation tags as indicated in all the write status entries 300 i .
  • the described operations of the storage controller driver 106 and the controller 112 in the storage devices 102 1 , 102 2 ...102 n may be implemented as a method, apparatus or computer readable storage medium using standard programming and/or engineering techniques to produce software, firmware, hardware, or any combination thereof.
  • the described operations may be implemented as code or logic maintained in a "computer readable storage medium", which may directly execute the functions or where a processor may read and execute the code from the computer storage readable medium.
  • the computer readable storage medium includes at least one of electronic circuitry, storage materials, inorganic materials, organic materials, biological materials, a casing, a housing, a coating, and hardware.
  • a computer readable storage medium may comprise, but is not limited to, a magnetic storage medium (e.g., hard disk drives, floppy disks, tape, etc.), optical storage (CD-ROMs, DVDs, optical disks, etc.), volatile and non-volatile memory devices (e.g., EEPROMs, ROMs, PROMs, RAMs, DRAMs, SRAMs, Flash Memory, firmware, programmable logic, etc.), Solid State Devices (SSD), etc.
  • the computer readable storage medium may further comprise digital logic implemented in a hardware device (e.g., an integrated circuit chip, a programmable logic device, a Programmable Gate Array (PGA), field-programmable gate array (FPGA), Application Specific Integrated Circuit (ASIC), etc.).
  • a hardware device e.g., an integrated circuit chip, a programmable logic device, a Programmable Gate Array (PGA), field-programmable gate array (FPGA), Application Specific Integrated Circuit (ASIC), etc.
  • the code implementing the described operations may be implemented in "transmission signals", where transmission signals may propagate through space or through a transmission media, such as an optical fiber, copper wire, etc.
  • the transmission signals in which the code or logic is encoded may further comprise a wireless signal, satellite transmission, radio waves, infrared signals, Bluetooth, etc.
  • the program code embedded on a computer readable storage medium may be transmitted as transmission signals from a transmitting station or computer to a receiving station or computer.
  • a computer readable storage medium is not comprised solely of transmission signals, but includes tangible components.
  • FIG. 15 illustrates an embodiment of a system 1500 in which the host system 100 may be implemented.
  • the system includes a processor 1504 that communicates over a bus 1506 with a volatile memory device 1508 in which programs, operands and parameters being executed are cached, such as the operating system 104 and storage controller driver 106, and a non-volatile storage device 1510, in which data and programs may be stored.
  • the processor 1504 may also communicate with Input/Output (I/O) devices 1512a, 1512b, which may comprise input devices, display devices, graphics cards, ports, network interfaces, etc.
  • the non-volatile storage device 1510 may be mounted to the system enclosure 1500, such as in a storage drive bay, or connected to the system 1500 through a port interface or over the network.
  • Example 1 is a computer readable storage device having computer program instructions executed in a host system in communication with a plurality of storage devices to cause operations, the operations comprising: determining a plurality of storage devices on which to write data for a write operation; generating a tag to uniquely identify the write operation; sending a write command to each of the determined storage devices including the tag and write data to cause each of the storage devices to write the write data at the storage device, wherein each of the storage devices maintains a copy of a previous version of the data to be updated by the write operation; and sending a revert command with the tag to one of the storage devices to cause the storage device to restore the copy of the previous version of the write data at the storage device.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Computer Security & Cryptography (AREA)
  • Quality & Reliability (AREA)
  • Techniques For Improving Reliability Of Storages (AREA)

Claims (11)

  1. Procédé exécuté dans un système hôte (100) en communication avec une pluralité de dispositifs de stockage (102), le procédé comprenant les étapes suivantes :
    déterminer une pluralité de dispositifs de stockage (102) sur lesquels écrire des données pour une opération d'écriture ;
    générer une étiquette (202 ; 302) pour identifier de manière unique l'opération d'écriture ;
    envoyer une commande d'écriture à chacun des dispositifs de stockage déterminés (102) incluant l'étiquette (202 ; 302) et des données d'écriture pour amener chacun des dispositifs de stockage (102) à écrire les données d'écriture sur le dispositif de stockage (102), où chacun des dispositifs de stockage (102) maintient une copie d'une version précédente des données devant être mises à jour par l'opération d'écriture ; et
    envoyer une commande de retour en arrière avec l'étiquette (202 ; 302) à l'un des dispositifs de stockage (102) pour amener le dispositif de stockage (102) à restaurer la copie de la version précédente des données sur le dispositif de stockage (102).
  2. Procédé selon la revendication 1, comprenant en outre les étapes suivantes :
    envoyer une commande d'ouverture à chacun des dispositifs de stockage (102) ayant l'étiquette (202 ;
    302), où la commande d'ouverture amène chacun des dispositifs de stockage (102) à maintenir la mise à jour de la copie de la version précédente des données par l'opération d'écriture ; et
    où la commande d'écriture est envoyée à chacun des dispositifs de stockage (102) en réponse à la réception d'un accusé de réception de la commande d'ouverture provenant de tous les dispositifs de stockage (102) qui ont envoyé la commande d'ouverture.
  3. Procédé selon la revendication 1, comprenant en outre l'étape suivante :
    en réponse à la détection d'une interruption, déterminer un état (401, 405) des dispositifs de stockage (102) qui n'ont pas accusé achèvement de la commande d'écriture, où la commande de retour en arrière est envoyée à au moins un des dispositifs de stockage (102) déterminés comme ayant achevé l'exécution de la commande d'écriture.
  4. Procédé selon la revendication 3, dans lequel la détermination de l'état (401, 405) des dispositifs de stockage (102) comprend d'envoyer une requête à chacun des dispositifs de stockage (102) ayant reçu la commande d'écriture qui n'ont pas accusé achèvement de la commande d'écriture, où la commande de retour en arrière est envoyée à chacun des dispositifs de stockage qui ont répondu à la commande d'écriture en indiquant que l'écriture a été achevée et à chacun des dispositifs de stockage qui ont répondu à la requête en indiquant que l'écriture a été achevée.
  5. Procédé selon la revendication 1, dans lequel les opérations comprennent en outre les étapes suivantes :
    placer en mémoire cache les données d'écriture envoyées avec les commandes d'écriture ;
    rejeter les données d'écriture placées en mémoire cache envoyées avec l'une des commandes d'écriture en réponse à la réception d'un accusé de réception indiquant que la commande d'écriture a achevé l'écriture des données ;
    en réponse à la détection d'une interruption, déterminer un état (401, 405) des dispositifs de stockage (102) qui ont reçu la commande d'écriture et qui n'ont pas accusé achèvement de l'écriture ; et
    renvoyer la commande d'écriture avec les données d'écriture placées en mémoire cache à chacun des dispositifs de stockage (102) ayant l'état indiquant que l'écriture n'a pas été achevée.
  6. Procédé exécuté par un dispositif de stockage (102) en communication avec un système hôte (100), où le dispositif de stockage (102) comprend un contrôleur qui, lorsqu'il est exécuté, exécute des opérations, les opérations comprenant les étapes suivantes :
    recevoir une commande d'écriture du système hôte (100) avec une étiquette (202 ; 302) identifiant de manière unique une opération d'écriture et des données d'écriture ;
    en réponse à la commande d'écriture, écrire les données d'écriture sur le dispositif de stockage (102) et maintenir une copie d'une version précédente des données mises à jour dans le cadre de la commande d'écriture ;
    recevoir une commande de retour en arrière avec l'étiquette identifiant l'opération d'écriture pour laquelle la commande d'écriture a été reçue ; et
    en réponse à la commande de retour en arrière, restaurer la copie de la version précédente des données mises à jour par les données d'écriture.
  7. Procédé selon la revendication 6, comprenant en outre les étapes suivantes :
    recevoir une commande de fermeture pour l'opération d'écriture ; et
    supprimer les informations sur l'opération d'écriture et la copie précédente des données pour l'opération d'écriture en réponse à la commande de fermeture de l'opération d'écriture ; et
    retourner au système hôte un accusé d'achèvement de l'exécution de la commande de fermeture.
  8. Procédé selon la revendication 6, comprenant en outre les étapes suivantes :
    recevoir une commande d'ouverture avec l'étiquette depuis le système hôte (100) pour l'opération d'écriture ;
    indiquer l'enregistrement d'intention pour l'étiquette de maintenir une copie d'une version précédente des données mises à jour dans le cadre de l'opération d'écriture identifiée par l'étiquette ; et
    renvoyer l'accusé de réception de la commande d'ouverture en réponse à l'indication de l'enregistrement d'intention, où la commande d'écriture est envoyée après renvoi d'un accusé de réception de la commande d'ouverture.
  9. Système (100) en communication avec une pluralité de dispositifs de stockage (102), comprenant des moyens adaptés pour exécuter les étapes du procédé de l'une quelconque des revendications 1 à 5.
  10. Dispositif de stockage (102) en communication avec un système hôte (100), comprenant des moyens adaptés pour exécuter les étapes du procédé de l'une quelconque des revendications 6 à 8.
  11. Stockage lisible par machine, comprenant des instructions lisibles par machine destinées, lorsqu'elles sont exécutées, à mettre en œuvre le procédé de l'une quelconque des revendications 1 à 8.
EP15867697.3A 2014-12-10 2015-11-12 Exécution d'une opération d'écriture atomique sur une pluralité de dispositifs de stockage Active EP3230846B1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US14/566,636 US9471448B2 (en) 2014-12-10 2014-12-10 Performing an atomic write operation across multiple storage devices
PCT/US2015/060478 WO2016094025A1 (fr) 2014-12-10 2015-11-12 Exécution d'une opération d'écriture atomique sur une pluralité de dispositifs de stockage

Publications (3)

Publication Number Publication Date
EP3230846A1 EP3230846A1 (fr) 2017-10-18
EP3230846A4 EP3230846A4 (fr) 2018-06-13
EP3230846B1 true EP3230846B1 (fr) 2019-12-25

Family

ID=56107928

Family Applications (1)

Application Number Title Priority Date Filing Date
EP15867697.3A Active EP3230846B1 (fr) 2014-12-10 2015-11-12 Exécution d'une opération d'écriture atomique sur une pluralité de dispositifs de stockage

Country Status (6)

Country Link
US (1) US9471448B2 (fr)
EP (1) EP3230846B1 (fr)
KR (1) KR102462708B1 (fr)
CN (1) CN107111453B (fr)
TW (1) TWI601008B (fr)
WO (1) WO2016094025A1 (fr)

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9760295B2 (en) * 2014-09-05 2017-09-12 Toshiba Memory Corporation Atomic rights in a distributed memory system
US9946607B2 (en) * 2015-03-04 2018-04-17 Sandisk Technologies Llc Systems and methods for storage error management
US10082985B2 (en) * 2015-03-27 2018-09-25 Pure Storage, Inc. Data striping across storage nodes that are assigned to multiple logical arrays
US9588857B2 (en) * 2015-06-22 2017-03-07 Lenovo Enterprise Solutions (Singapore) Pte. Ltd. Raid logical stripe backup to non-volatile memory in response to raid storage device media errors
US10120751B2 (en) * 2015-09-25 2018-11-06 Intel Corporation Techniques to recover data using exclusive OR (XOR) parity information
US9921914B2 (en) 2015-11-03 2018-03-20 Intel Corporation Redundant array of independent disks (RAID) write hole solutions
US9910786B2 (en) 2015-11-03 2018-03-06 Intel Corporation Efficient redundant array of independent disks (RAID) write hole solutions
US10379838B1 (en) 2017-06-28 2019-08-13 Amazon Technologies, Inc. Update and rollback of code and API versions
US10725763B1 (en) * 2017-06-28 2020-07-28 Amazon Technologies, Inc. Update and rollback of configurations in a cloud-based architecture
US10552061B2 (en) * 2017-08-08 2020-02-04 International Business Machines Corporation Providing preferential access to a metadata track in two track writes
EP3726364B1 (fr) * 2017-12-28 2022-07-27 Huawei Technologies Co., Ltd. Procédé d'écriture de données et ensemble de disques durs électroniques
CN110825309B (zh) * 2018-08-08 2021-06-29 华为技术有限公司 数据读取方法、装置及系统、分布式系统
CN108918928B (zh) * 2018-09-11 2020-11-10 广东石油化工学院 一种负荷分解中功率信号自适应重构方法
US10891232B2 (en) * 2019-03-21 2021-01-12 International Business Machines Corporation Page-based memory operation with hardware initiated secure storage key update
US11687471B2 (en) * 2020-03-27 2023-06-27 Sk Hynix Nand Product Solutions Corp. Solid state drive with external software execution to effect internal solid-state drive operations
CN111897495B (zh) * 2020-07-28 2023-07-04 深圳忆联信息系统有限公司 提高ssd写性能的实现方法、装置、计算机设备及存储介质
US20230136664A1 (en) * 2021-11-02 2023-05-04 SK Hynix Inc. Storage device and electronic device

Family Cites Families (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5774643A (en) * 1995-10-13 1998-06-30 Digital Equipment Corporation Enhanced raid write hole protection and recovery
JP3592640B2 (ja) * 2001-01-09 2004-11-24 株式会社東芝 ディスク制御システムおよびディスク制御方法
US6523087B2 (en) * 2001-03-06 2003-02-18 Chaparral Network Storage, Inc. Utilizing parity caching and parity logging while closing the RAID5 write hole
US6857053B2 (en) 2002-04-10 2005-02-15 International Business Machines Corporation Method, system, and program for backing up objects by creating groups of objects
US7401191B1 (en) * 2004-05-28 2008-07-15 Ncr Corp. System and method for performing disk write operations by writing to a data depot prior to an in-place write
US7636814B1 (en) 2005-04-28 2009-12-22 Symantec Operating Corporation System and method for asynchronous reads of old data blocks updated through a write-back cache
US7613743B1 (en) 2005-06-10 2009-11-03 Apple Inc. Methods and apparatuses for data protection
KR101340176B1 (ko) * 2005-08-25 2013-12-10 실리콘 이미지, 인크. 스마트 스케일러블 스토리지 스위치 아키텍처
US8443153B1 (en) * 2010-01-06 2013-05-14 Netapp, Inc. Dynamic balancing of performance with block sharing in a storage system
EP2598996B1 (fr) * 2010-07-28 2019-07-10 SanDisk Technologies LLC Appareil, système et procédé d'opérations de stockage conditionnel et atomique
US9569320B2 (en) * 2010-12-01 2017-02-14 Seagate Technology Llc Non-volatile memory program failure recovery via redundant arrays
US8732518B2 (en) 2011-04-13 2014-05-20 Netapp, Inc. Reliability based data allocation and recovery in a storage system
US20130198446A1 (en) * 2012-01-30 2013-08-01 Infinidat Ltd. Storage system for atomic write of one or more commands
US9116793B2 (en) 2012-06-12 2015-08-25 International Business Machines Corporation Maintaining versions of data in solid state memory
US9395924B2 (en) * 2013-01-22 2016-07-19 Seagate Technology Llc Management of and region selection for writes to non-volatile memory
KR20140128819A (ko) * 2013-04-29 2014-11-06 삼성전자주식회사 아토믹 라이트 방법
EP2879040B1 (fr) * 2013-10-18 2017-03-01 Huawei Technologies Co., Ltd. Procédé de stockage de données, appareil de stockage de données, et dispositif de stockage
US9274720B1 (en) * 2014-09-15 2016-03-01 E8 Storage Systems Ltd. Distributed RAID over shared multi-queued storage devices

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
None *

Also Published As

Publication number Publication date
WO2016094025A1 (fr) 2016-06-16
KR20170092537A (ko) 2017-08-11
US9471448B2 (en) 2016-10-18
CN107111453A (zh) 2017-08-29
CN107111453B (zh) 2020-11-06
US20160170850A1 (en) 2016-06-16
TW201633127A (zh) 2016-09-16
EP3230846A1 (fr) 2017-10-18
EP3230846A4 (fr) 2018-06-13
TWI601008B (zh) 2017-10-01
KR102462708B1 (ko) 2022-11-04

Similar Documents

Publication Publication Date Title
EP3230846B1 (fr) Exécution d'une opération d'écriture atomique sur une pluralité de dispositifs de stockage
CN106471478B (zh) 用于在非易失性数据存储设备内以原子的方式执行多个写事务的设备控制器和方法
US10191812B2 (en) Recovery mechanism for low latency metadata log
US9218278B2 (en) Auto-commit memory
US9785575B2 (en) Optimizing thin provisioning in a data storage system through selective use of multiple grain sizes
US9684591B2 (en) Storage system and storage apparatus
US9652164B2 (en) Solid-state mass storage device and method for processing forced unit access write commands
EP3633517B1 (fr) Dispositif de stockage et procédé de fonctionnement d'un dispositif de stockage
US20190324859A1 (en) Method and Apparatus for Restoring Data after Power Failure for An Open-Channel Solid State Drive
US20190369892A1 (en) Method and Apparatus for Facilitating a Trim Process Using Auxiliary Tables
US11693768B2 (en) Power loss data protection in a memory sub-system
US10521148B2 (en) Data storage device backup
US11630778B2 (en) Efficient processing of commands in a memory sub-system
CN112148533A (zh) 将数据存储在存储器系统的mlc区域中的设备和方法
CN110968522A (zh) 存储器系统及其操作方法、包括存储器系统的数据库系统
US10031689B2 (en) Stream management for storage devices
KR102094475B1 (ko) 멀티-트랜잭션의 아토믹 라이트 방법
WO2016032955A2 (fr) Systèmes de stockage prenant en charge la nvram
CN116774913A (zh) 改善混合读/写操作期间的读取时延的存储系统和方法
KR20220086934A (ko) 비휘발성 메모리 시스템의 저널링 제어 장치 및 방법
CN115248748A (zh) 执行元数据管理的存储装置及操作该存储装置的方法
CN115048042A (zh) 启用针对持久性存储器的存储器存取事务
WO2015196416A1 (fr) Procédé et dispositif de stockage de données et mémoire non volatile

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20170510

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
A4 Supplementary search report drawn up and despatched

Effective date: 20180514

RIC1 Information provided on ipc code assigned before grant

Ipc: G06F 11/14 20060101ALI20180508BHEP

Ipc: G06F 3/06 20060101AFI20180508BHEP

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: GRANT OF PATENT IS INTENDED

INTG Intention to grant announced

Effective date: 20190531

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE PATENT HAS BEEN GRANTED

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 602015044446

Country of ref document: DE

REG Reference to a national code

Ref country code: AT

Ref legal event code: REF

Ref document number: 1217863

Country of ref document: AT

Kind code of ref document: T

Effective date: 20200115

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: NL

Ref legal event code: FP

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: BG

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20200325

Ref country code: FI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: LV

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: SE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: NO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20200325

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20200326

Ref country code: LT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

REG Reference to a national code

Ref country code: LT

Ref legal event code: MG4D

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: HR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: RS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: AL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: EE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: CZ

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: RO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: PT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20200520

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SM

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20200425

Ref country code: SK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

REG Reference to a national code

Ref country code: DE

Ref legal event code: R097

Ref document number: 602015044446

Country of ref document: DE

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: ES

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

REG Reference to a national code

Ref country code: AT

Ref legal event code: MK05

Ref document number: 1217863

Country of ref document: AT

Kind code of ref document: T

Effective date: 20191225

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

26N No opposition filed

Effective date: 20200928

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: AT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: IT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: PL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MC

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

REG Reference to a national code

Ref country code: CH

Ref legal event code: PL

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: LU

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20201112

REG Reference to a national code

Ref country code: BE

Ref legal event code: MM

Effective date: 20201130

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CH

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20201130

Ref country code: LI

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20201130

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20201130

Ref country code: IE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20201112

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: TR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: MT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

Ref country code: CY

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191225

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: BE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20201130

P01 Opt-out of the competence of the unified patent court (upc) registered

Effective date: 20230518

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: NL

Payment date: 20230915

Year of fee payment: 9

Ref country code: GB

Payment date: 20230914

Year of fee payment: 9

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: DE

Payment date: 20230919

Year of fee payment: 9